Oscilloscopes are one of the more common examples of electrical test and measurement equipment. This class of devices makes primary measurements of time varying signals. In most instances, these measurements are of the signal's voltage as a function of time, even though the detected voltage may be indicative of some other measurement of interest such as electrical current through a resistive element or temperature in the case of a thermistor, in two arbitrary examples. The oscilloscope is useful because it enables the visualization of the time varying character of signals, using a vertical axis representing level and horizontal axis representing time.
The digital storage oscilloscope (DSO) is a subclass of oscilloscopes in which the time varying nature of the sampled signals is represented digitally within the device. The main advantage is that non-simultaneous signal events can be stored in the device for subsequent comparison. Additionally, parameters can be derived from the digital data of the primary measurements, such as statistical features of the signals.
In the typical implementation, the DSO works by waiting for the satisfaction of some trigger condition. When the trigger event is received, the primary measurements of the voltage, for example, of the signal are made, and the resulting measurement data are stored in a waveform memory. Successive positions in the waveform memory hold the digitized level of the signal at increasing time delays from the triggering event. Under current technology, DSOs can capture signals at a rate of up to 8 gigasamples/second (GS/s) with waveform memories of up to 16 million storage locations.
Where oscilloscopes are time domain instruments, spectrum analyzers make primary measurements in the frequency domain. In the typical configuration, these devices plot the magnitude of the signal energy as a function of frequency--signal magnitude or level being on the vertical axis with the frequency on the horizontal axis. These devices typically operate by scanning a very narrow notch-bandpass filter across the frequency spectrum of interest and measuring energy in the frequency bins. In this way, spectrum analyzers are useful in identifying the spectral distribution of a given signal.
For certain signal analysis problems, however, oscilloscopes and spectrum analyzers are not well suited to the task. For example, when trying to isolate infrequent anomalies in a signal such as that required for digital/analog circuit analysis/debugging or when trying to identify trends in signals such as when analyzing modulated signals, both oscilloscopes and spectrum analyzers are less useful. Since spectrum analyzers operate by scanning filters across the spectrum, any short term changes in the signals are lost; and in most oscilloscopes, such infrequent anomalies will scroll by on the display at a rate that is too fast for the operator to analyze. Only DSOs retain the relevant information on the signals, but in order to analyze it, the operator must scan through long arrays of data to find events that may not be readily apparent from the primary measurements alone.
In order to fill this gap, electronic counters such as modulation domain analyzers have been developed. These devices operate by performing primary measurements of the signal, i.e., detecting the signal's crossing time of a set threshold, and then generating plots of parameters derived from the primary measurements, such as the signal's frequency, phase, or time interval as a function of time. For example, the frequency/phase versus time analysis are very useful in analyzing frequency-shift-key and phase-shift-key, respectively, modulated transmissions; time interval analysis is useful for analyzing pulse-width modulated signals.